Supplemental Material can be found at: http://jn.nutrition.org/content/suppl/2014/03/12/jn.113.19026 4.DCSupplemental.html
The Journal of Nutrition Nutritional Immunology
Dietary Bovine Lactoferrin Alters Mucosal and Systemic Immune Cell Responses in Neonatal Piglets1–3 Sarah S. Comstock,4 Elizabeth A. Reznikov,5 Nikhat Contractor,6 and Sharon M. Donovan4,5*
Abstract Lactoferrin (LF) is a multifunctional immune protein found at high concentrations in human milk. Herein, the effect of dietary bovine LF (bLF) on mucosal and systemic immune development was investigated. Colostrum-deprived piglets were fed formula containing 130 [control (Ctrl)], 367 (LF1), or 1300 (LF3) mg of bLF/(kg body weight d). To provide passive immunity, sow serum was provided orally during the first 36 h of life. Blood, spleen, mesenteric lymph node (MLN), and ascending colon (Asc) contents were collected on day 7 (n = 10–14/group) and day 14 (n = 10–12/group). Immune cell populations were quantified by flow cytometry and immunoglobulins (Igs) were measured by ELISA. Additionally, immune cells were isolated from spleen and MLNs (n = 7/group) on day 7 and stimulated ex vivo with phytohemagglutinin or lipopolysaccharide (LPS) 6 LF for 72 h. Secreted cytokine concentrations were quantified by multiplex assay. Lymphocyte populations [cluster determinant (CD)4, CD8, and natural killer cells] developed normally and were unaffected by dietary bLF. LF3 piglets tended to have 1.4 to 2 times more serum IgG than Ctrl piglets (P = 0.07) or LF1 piglets (P = 0.03), but IgA in Asc contents was unaffected by bLF. Asc IgA was 4 times higher on day 14 than day 7. Spleen cells from LF3 piglets produced 2 times more interleukin (IL)-10 and tumor necrosis factor (TNF)-a ex vivo than those from Ctrl or LF1 piglets. MLN cells from LF1 and LF3 piglets produced 40% more IL-10 and tended to produce 40% more IL-6 (P = 0.05) than those from Ctrl piglets. However, ex vivo bLF did not affect the cytokine response of spleen or MLN cells to LPS. In summary, dietary bLF alters the capacity of MLN and spleen immune cells to respond to stimulation, supporting a role for LF in the initiation of protective immune responses in these immunologically challenged neonates. J. Nutr. 144: 525–532, 2014.
Introduction Clinical and epidemiologic evidence suggest that human infants fed formula have altered immune responsiveness compared with those fed human milk (1,2). It is well known that human milk and formula differ in composition (3,4). One protein whose concentration markedly differs between human milk and infant formula is lactoferrin (LF)7. Human colostrum and mature milk contain ;7000 and 2500 mg LF/L, respectively (5,6). In contrast, 1
Supported by Wyeth Nutrition, a Nestle´ Business, King of Prussia, PA. Author disclosures: N. Contractor was an employee of Wyeth Nutrition at the time of this study. S. M. Donovan and S. S. Comstock have consulted for Wyeth Nutrition. E. A. Reznikov, no conflicts of interest. 3 Supplemental Methods, Supplemental Table 1, and Supplemental Figures 1–5 are available from the ‘‘Online Supporting Material’’ link in the online posting of the article and from the same link in the online table of contents at http://jn. nutrition.org. 7 Abbreviations used: Asc, ascending colon; bLF, bovine lactoferrin; BW, body weight; CD, cluster determinant; Ctrl, control, 130 mg bLF/(kg BWd); IPP, ileal PeyerÕs patch; LF, lactoferrin; LF1, 367 mg bLF/(kg BWd); LF3, 1300 mg bLF/(kg BWd); LFR, lactoferrin receptor; MLN, mesenteric lymph node; PBMC, peripheral blood mononuclear cell; PHA, phytohemagglutinin; PreLPS, LF pretreated with LPS. * To whom correspondence should be addressed. E-mail: sdonovan@illinois. edu. 2
bovine milk contains between 50 and 300 mg LF/L (7). This low concentration in bovine milk results in cow milk–based infant formulas containing small quantities of LF (8). LF is an iron-binding protein that has immune-modulating effects (9), antimicrobial functionality (10), and antioxidant activity (8). LF has direct bacteriocidal effects, but may also directly interact with the host immune system through interactions with pathogen-associated molecules, such as CpG-rich DNA or LPS (11–14). Additional immune-modulating effects of LF may be receptor-mediated as activated lymphocytes express LF receptors (15–17). As a product of neutrophil granule release, LF likely plays an important role in the local microenvironment of an immune response (18). Orally administered bovine lactoferrin (bLF) has been demonstrated to influence human immunity (19,20). Thus, milk-derived LF may be important in neonatal immune response and development. The neonatal piglet is a well-accepted model for human infant gastrointestinal function, growth, and metabolism. Sow milk contains LF (21), the LF receptors are present in the porcine intestine (22), and orally administered LF is absorbed by the neonatal piglet (23). For these reasons, we used the neonatal piglet to study how dietary bLF influences immune cell populations and
ã 2014 American Society for Nutrition. Manuscript received December 24, 2013. Initial review completed January 4, 2014. Revision accepted January 28, 2014. First published online February 19, 2014; doi:10.3945/jn.113.190264.
525
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
4 Food Science and Human Nutrition and 5Nutritional Sciences, University of Illinois, Urbana, IL; and 6Wyeth Nutrition, a Nestle´ Business, King of Prussia, PA
responsiveness in early life. We hypothesized that dietary bLF would improve the immune development of formula-fed neonates. We anticipated that bLF supplementation would increase cytotoxic T-cell numbers, prime stimulated immune cells to produce more cytokines, and increase immunoglobulin production.
Materials and Methods
Sample collection. On day 7 (n = 36) or 14 (n = 32) postpartum, piglets were sedated with an intramuscular injection of Telazol (Tiletamine HCl and Zolazepam HCl, 3.5 mg/kg BW each; Pfizer Animal Health, Fort Dodge, IA). After sedation, blood was collected by cardiac puncture into EDTA-laced vials (BD Biosciences, Franklin Lakes, NJ) for isolation of mononuclear cells. Piglets were then killed by an intravenous injection of sodium pentobarbital (72 mg/kg BW; Fatal Plus; Vortech Pharmaceuticals, Dearborn, MI) as described previously (25). Mesenteric lymph nodes (MLNs), spleens, and ileal PeyerÕs patches (IPPs) were quickly dissected and ascending colon (Asc) contents were collected. Isolation of mononuclear cells from peripheral blood and cells from immune tissues. Peripheral blood mononuclear cells (PBMCs) were isolated as previously described (26). Supplemental Methods provides detailed isolation procedures. Briefly, spleens, MLNs, and IPPs were washed. IPPs were incubated in DTT/HBSS and then with EDTA/HBSS. Spleens, MLNs, and IPPs were rinsed and single cell suspensions were made through a combination of enzymatic and mechanical digestion. Resulting cell suspensions were treated as described (25). Cell stimulation. Cells from 7-d-old piglets were plated (96-well plate, 2 3 105 cells/well) in a final volume of 200mL complete Roswell Park Memorial Institute at 37°C under 5% CO2. Stimulants were added immediately (n = 2 wells per sample per stimulant). Cells were stimulated with LPS, a toll-like receptor 4 agonist, (2 mg/mL); phytohemagglutinin (PHA), a lectin that binds and crosslinks the T-cell receptor, (2.5 mg/ml); or bLF (50 mg/mL). In some cases bLF was added with polymixin B (10 mg/ mL) to control for potential endotoxin in the bLF solution. Additionally, the following mixtures were used to stimulate cells: LPS + LF, LF + polymixin B, and LF pretreated with LPS (PreLPS) for 1 h prior to addition to cells (PreLPS + LF). Cell culture supernatants were collected 72 h after initiation of culture and frozen at 280°C until analyzed. Cytokine production. Supernatants were analyzed for IL-6, IL-10, IFNg, IL-12p40, IL-4, and TNF-a by Panomics using a Luminex-based assay 526
Comstock et al.
Phenotypic identification of cells. Cells were resuspended in flow staining buffer (PBS, 1% BSA, 0.1% sodium azide). Cell phenotypes were determined by flow cytometry as previously described (26). To differentiate T and NK cells, cells were stained with anti-pig cluster determinant (CD) 3, anti-pig CD4, and anti-pig CD8 (27). B lymphocytes were identified by anti-pig CD21 and anti-pig swine leukocyte antigen class II DR. Polymorphonuclear cells were identified by anti-pig CD163, anti-pig CD172a, and anti-pig swine workshop cluster 8 (28). Detailed information is included in Supplemental Methods. Staining was assessed by a LSRII flow cytometer (BD Biosciences) and analyzed using FlowJo 7.0 (FlowJo). Immunglobulin and bLF concentrations. Serum IgG, serum bLF, and Asc content IgA were determined using ELISA antibody sets (Bethyl Labs). The manufacturerÕs instructions were followed with the exception that PBS was used as the basis for assay buffers. Prior to analysis, Asc contents were extracted (1:3) using a PBS/protease inhibitor solution (Roche). IgA concentrations are reported as mg IgA per mg total protein in the extract. Amounts of bLF in diets were analyzed by semiquantitative Western blot at Wyeth Nutrition. Supplemental Methods provides details. Statistical analyses. Data were analyzed using SAS 9.2. Detailed information about statistical analyses is in Supplemental Methods. Proc general linear model was used in most cases with day, diet, and day diet as factors, or diet, stimulant, and diet stimulant as factors. Post hoc testing was least-squares means. For IgG, sow had an effect in the Proc general linear model, so Proc MIXED was used with fixed effects of diet and day and random effects including sow and replicate. If data were not normally distributed, they were log transformed to be normally distributed or, when log transformation failed, analyzed by nonparametric tests. LF intake was analyzed by WilcoxonÕs rank-sum test to compare days and by Kruskal-Wallis 1-factor ANOVA to compare diets. Serum bLF concentration was analyzed using a nonparametric Friedman 2-factor ANOVA. For survival analysis, the LIFETEST procedure was used to generate Kaplan-Meier survival curves and to compare survival curves among the dietary treatment groups. Differences were considered significant at P < 0.05 and trends at P < 0.10. Data are expressed as means 6 SEMs.
Results Bovine lactoferrin intake During the first 7 d, piglets consumed 103 6 13 (Ctrl), 321 6 26 (LF1), and 1170 6 39 (LF3) mg bLF/(kg BW d), respectively. Throughout the 14-d study, piglets consumed 114 6 14 (Ctrl), 413 6 27 (LF1), and 1270 6 37 (LF3) mg bLF/(kg BWd), respectively. Thus, 7-d-old and 14-d-old piglets were consuming similar amounts of bLF on a per-kg BW basis, but bLF intake differed across diets (P < 0.0001). However, on an absolute basis, piglets analyzed at day 7 consumed 256 6 25 (Ctrl), 836 6 111 (LF1), and 3130 6 201 mg bLF/d (LF3) during the 3 days prior to sampling. Piglets analyzed at day 14 consumed 496 6 82 (Ctrl), 1790 6 159 (LF1), and 5130 6 335 mg bLF/d (LF3) during the 3 d prior to sampling. Thus, the absolute dose of bLF was much higher just prior to sampling on day 14 than on day 7 within the Ctrl (P = 0.002), LF1 (P = 0.0003), and LF3 (P = 0.0002) groups. The absolute dose of LF also differed across diets (P < 0.0001). Piglet growth and health status Birth weights were similar (1.6 6 0.04 kg). Neither absolute body weight nor weight gain of piglets was affected by dietary bLF. On day 7, piglets weighed 2.5 6 0.07 kg. On day 14, piglets weighed 4.2 6 0.11 kg. Spleen weight was unaffected by dietary
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
Piglets, housing, and care. Pregnant sows (n = 10) were supplied by the Swine Research Center at the University of Illinois, UrbanaChampaign. Vaginally delivered piglets (n = 92; n = 2–17 per sow) were removed from the sow immediately, prior to consumption of colostrum, and were washed with a povidone-iodine microbicidal solution (Henry Schein Animal Health). Because piglets receive no maternally transferred antibodies prenatally, piglets were administered pregnant sow serum at birth [4 mL/kg body weight (BW)] and 5 mL/kg BW at 8 h, 22 h, and 36 h postnatal age by oro-gastric gavage to provide passive immunity. Pigs were fed a sow milk replacer formula (Animix; Supplemental Table 1) at a rate of 360 mL/(kg BW d) divided equally into 22 feedings. Piglets were assigned to 1 of 3 dietary treatment groups at birth: control formula [(Ctrl), n = 34], or formula with added bLF at either 1000 mg bLF/L [367 mg bLF/(kg BW d) (LF1), n = 29)] or 3620 mg bLF/L [1300 mg/(kg BW d) (LF3), n = 29]. Group (dietday) enrollment was balanced by sex, sow, and birth weight. The control diet contained innate amounts of 360 mg bLF/L [130 mg/(kg BW d)]. The 2 LF concentrations were selected to reflect the dose of LF consumed by breastfed human infants (LF1) or 5 times that dose (LF3) on a mg/kg BW basis. This dose assumed that human infants (;5.5 kg BW) consume about 1 L of human milk/d with an LF concentration of 1850 mg/L (24). The bLF preparation was 97% pure LF (DMV). Piglets were individually housed as described (25). All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of Illinois. Institutional and national guidelines for the care and use of animals were followed.
specific for porcine cytokines. When the value for a sample fell below the limit of detection of the assay, values were set to 0.
Bovine lactoferrin concentration in serum To determine whether dietary LF was being absorbed, bLF concentration in piglet serum was measured. Most samples were below the linear range (500–7.8 mg/L) of the assay. For piglets with quantifiable serum bLF (1 day 7 LF3, 3 day 14 LF3, and 1 day 14 LF1), the concentration was 16.1 6 3.6 mg/L serum. Because few samples contained concentrations of bLF within the linear range of the assay, absorbance values above assay background were used for statistical analysis by ranks. LF3 piglets had higher serum bLF concentrations than Ctrl or LF1 piglets (P < 0.0001, Supplemental Fig. 2). LF1 piglets had higher serum bLF concentrations than Ctrl piglets (P = 0.0001). Immunoglobulin concentrations Total serum IgG concentrations were similar on day 7 and day 14. However, LF3 piglets had higher total serum IgG than LF1 piglets (P = 0.03) and tended to have higher total serum IgG than
Ctrl piglets (P = 0.07, Fig. 1A). In contrast, diet did not affect the IgA concentrations in Asc contents, but total Asc IgA concentrations were higher on day 14 than day 7 (P < 0.0001, Fig. 1B). T-cell populations T helper cells decreased, cytotoxic T cells increased, and the CD4 to CD8 T-cell ratio decreased in PBMC, MLN, spleen, and IPP as piglets aged (Table 1). Dietary bLF did not affect T-cell populations (Supplemental Fig. 3A–C). Neither time nor diet affected the proportion of double positive CD4+CD8+ (a mature effector memory cell population in pigs) within PBMC, MLN, spleen, or IPP (Supplemental Fig. 3D). Other immune cell populations Immune cell populations were affected by age but not by dietary bLF. The NK (CD3-CD4-CD8+) cell populations in PBMC, MLN, and spleen increased, whereas PBMC and IPP B-cell populations decreased with age (Table 2). Total major histocompatibility complex II+ cell populations decreased in PBMC, MLN, and IPP, but increased in spleen, with age (Table 2). Polymorphonuclear cells in the MLN and IPP decreased with age (Table 2). Dietary bLF did not affect immune cell populations (Supplemental Fig. 4A–D). Cytokine production PHA-induced. Splenocytes from 7-d-old piglets fed 1300 mg LF/ (kg BW d) (LF3) produced more IFN-g than cells from the spleen of Ctrl piglets (Fig. 2). IFN-g production by cells isolated from the MLNs of Ctrl and LF3 piglets was similar, and dietary LF did not affect IL-6 or IL-10 production in response to PHA stimulation of cells isolated from MLNs or spleen (Supplemental Fig. 5). LPS-induced. Dietary, but not ex vivo, bLF altered cytokine production by spleen and MLN mononuclear cells. Ex vivo bLF did not affect cellular cytokine production in response to LPS stimulation (Figs. 3 and 4). This was true whether the LPS and bLF were pre-incubated for 1 hour prior to addition to cell culture (PreLPS + LF) or were added to cell culture simultaneously (LPS + LF). Dietary LF substantially affected cytokine production by MLNs (Fig. 3) and spleen (Fig. 4) cells from 7-dold piglets. MLN cells from LF1 piglets produced more IFN-g than those isolated from Ctrl piglets (Fig. 3A). MLN cells from LF1 piglets produced more IL-6 and more IL-10 than those isolated from Ctrl piglets (Fig. 3B, C). LF3 piglet MLN cells tended to produce more IL-6 (P = 0.05) and did produce more IL-10 than cells isolated from Ctrl piglets (Fig. 3B, C). MLN production of TNF-a was not affected by dietary bLF (Fig. 3D). Splenocytes from either LF1 or LF3 piglets produced more IFN-g than those from Ctrl piglets (Fig. 4A). IL-6 production by splenocytes was not affected by dietary LF (Fig. 4B). Only splenocytes from LF3 piglets produced more IL-10 or TNF-a than Ctrl piglets (Fig. 4C, D).
Discussion
FIGURE 1 Serum IgG (A) and Asc contents IgA (B) in piglets fed Ctrl, LF1, or LF3 for 7 or 14 d. Values are means 6 SEMs, n = 10–13. Asc, ascending colon; bLF, bovine lactoferrin; BW, body weight; Ctrl, control, 130 mg bLF/(kg BW d); D, day; LF1, 367 mg bLF/(kg BW d); LF3, 1300 mg bLF/(kg BW d); T, treatment.
The key finding from this study was an improvement in the immune response of colostrum-deprived neonatal piglets fed high doses of bLF, including in systemic and mucosal immune cells capable of producing higher quantities of cytokines than in piglets fed formula containing lower quantities of bLF. Although no changes in immune cell population sizes were observed, the trend toward (P = 0.07) increased serum IgG concentrations, substantially enhanced cytokine production, and lower mortality observed in piglets fed bLF support a role for LF in the initiation of protective immune responses in these immunologically challenged neonates. Dietary lactoferrin alters immune response
527
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
bLF but was affected by age, with the spleens of 14-d-old piglets (2.6 6 0.2 g/kg BW) being larger than those of 7-d-old piglets (2.0 6 0.1 g/kg BW) (P = 0.003). Liver weight (41.5 6 0.6 g/kg BW) was unaffected by dietary bLF and age. Because piglets were colostrum-deprived, there was some mortality during the study (29). Pairwise (LF3 vs. Ctrl, LF1 vs. Ctrl) comparisons were conducted. Dietary bLF at 367 mg/(kg BW d) (LF1) was not sufficient to alter survival (Supplemental Fig. 1A). However, piglets fed the highest bLF dose [1300 mg/(kg BW d), LF3] were more likely to survive than those consuming the control diet (P = 0.03) (Supplemental Fig. 1B). When survival of piglets in all 3 treatment groups was compared simultaneously, diet tended (P = 0.08) to affect survival.
TABLE 1
T-cell populations in the PBMCs, MLNs, spleen, and IPPs of 7- and 14-d-old piglets1 Day 7
P value
77.3 80.8 80.4 90.9
6 1.2 6 0.8 6 1.8 6 1.3
69.5 6 76.8 6 73.6 6 79.7 6
1.6 1.1 1.7 2.5
0.0011 0.004 0.013 0.0001
5.8 4.5 4.8 1.5
6 0.5 6 0.7 6 0.8 6 0.4
9.8 6 8.0 6 6.1 6 1.8 6
0.7 0.8 0.6 0.3
,0.0001 0.0002 0.006 0.003
19.2 40.7 46.9 314
6 2.2 6 5.3 6 6.2 6 44
9.1 6 17.1 6 20.2 6 97.3 6
1.1 4.3 3.4 15.9
,0.0001 0.0016 0.0007 0.002
5.7 11 5.9 2.3
6 0.4 61 6 0.4 6 0.2
5.7 6 11 6 5.9 6 3.6 6
0.6 1 0.3 0.5
0.89 0.71 0.96 0.14
1 T-cell populations expressed as % total CD3+ events. Values are means 6 SEMs, n = 34–38. CD, cluster determinant; IPP, ileal PeyerÕs patch; MLN, mesenteric lymph node; PBMC, peripheral blood mononuclear cell.
In previous studies, bLF was shown to increase animal growth (30–34). In our study, dietary bLF did not affect the growth of colostrum-deprived piglets. This is likely because bLF was incorporated into the diets at the time of formulation, resulting in diets that were isocaloric and isonitrogenous, whereas bLF was provided as a supplement in previous studies (32,34). The addition of high-quality protein, rather than a specific growthpromoting action of bLF, could be the reason for improved growth in some studies. Some piglets died during the study, which is to be expected because they were colostrum-deprived and not maintained in a specific pathogen-free facility. Thus, the time period between birth and gavage with sow serum, albeit short, left piglets vulnerable to potential infection. Serum IgG concentrations of piglets in this study were lower than those in piglets consuming maternal colostrum and milk (35) and were about half (36) to one-fourth the amounts (37) previously reported for colostrumdeprived piglets. These discrepancies are likely due to the specific pool of sow serum used to transfer passive immunity to the piglets in this study. Low serum Ig concentrations likely contributed to mortality. A recent study suggests that serum IgG concentrations below 10 mg/mL increase piglet mortality (29). Unlike piglets, human infants receive maternal Ig through placental transfer (38,39), with preterm infants receiving less maternal Ig (39,40). As a result, colostrum-deprived human infants are less vulnerable at birth than colostrum-deprived piglets. However, it is well recognized that even term infants are somewhat immunocompromised because of their inexperienced and underdeveloped immune systems (41–43). Additionally, the similarity of our environmental conditions to those in the real world improves the translation of this research to humans. Thus, the increased risk of piglet mortality is justified because the conditions that increase mortality improve the applicability of the research to human infants. Several additional aspects also suggest these results would apply to humans. Piglets are outbred mammals known to be excellent models for human gastrointestinal and immune devel528
Comstock et al.
opment (44–46). Specifically, T-, B-, and NK-cell development observed in the current study was similar to what is observed in human immune development (2,47–49) and what was previously observed in pigs (25). These outbred piglets were maintained in an environment containing a wide variety of microbes typical of those found in a normal room, i.e., piglets were not maintained in a specific pathogen-free environment. All piglets began the TABLE 2 Other immune cell populations in the tissues of 7and 14-d-old piglets1 Day 7 NK cells,2 CD3-CD4-CD8+ PBMC MLN Spleen IPP B-cells,3 CD21+MHCII+ PBMC MLN Spleen IPP MHCII+ cells3 PBMC MLN Spleen IPP PMN,4 SWC8+CD172a+CD1632 MLN IPP
4.2 6 1.1 6 3.5 6 0.6 6
Day 14
P value
0.8 0.2 0.9 0.1
8.8 2.1 3.9 0.5
6 1.3 6 0.3 6 0.4 6 0.1
0.0120 0.0002 0.0015 0.42
13.0 32.8 21.7 77.3
6 6 6 6
1.2 2.1 1.7 3.2
9.2 29.8 23.0 65.4
6 0.9 6 2.2 6 1.4 6 2.4
0.0320 0.33 0.53 0.0050
70.7 40.5 41.1 77.4
6 6 6 6
3.1 2.0 2.9 3.2
58.1 35.3 49.6 65.4
6 3.0 6 2.2 6 2.7 6 2.4
0.0048 0.09 0.0389 0.0086
27.5 6 2.3 28.5 6 3.1
0.0040 ,0.0001
40.1 6 3.4 54.8 6 4.5
Values are means 6 SEMs, n = 34–38. CD, cluster determinant; IPP, ileal PeyerÕs patch; MHC, major histocompatibility complex; MLN, mesenteric lymph node; PBMC, peripheral blood mononuclear cell; PMN, polymorphonuclear cell; SWC, swine workshop cluster. 2 Expressed as % CD3–events. 3 Expressed as % lymphocytes. 4 Expressed as % CD163–events. 1
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
T-helper cells, CD3+CD4+CD82 PBMC MLN Spleen IPP Cytotoxic T-cells, CD3+CD4-CD8+ PBMC MLN Spleen IPP Helper/cytotoxic ratio PBMC MLN Spleen IPP Mature effector memory T-cells, CD3+CD4+CD8+ PBMC MLN Spleen IPP
Day 14
indicates a general priming of the IL-10 response in piglets fed bLF. Because IL-10 is a key mediator of intestinal homeostasis as well as immune regulation and is expressed by both innate and adaptive immune cells (66), increased production of this cytokine
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
FIGURE 2 IFN-g production by PHA-stimulated spleen cells isolated from 7-d-old LF3 and Ctrl piglets. Values are means 6 SEMs, n = 4 per diet. bLF, bovine lactoferrin; BW, body weight; Ctrl, control, 130 mg bLF/(kg BW d); LF3, 1300 mg bLF/(kg BW d); PHA, phytohemagglutinin; S, stimulant; T, treatment; Ust, unstimulated.
study with normalized passive immunity achieved through oral gavage with pooled sow serum. The decreased mortality in piglets fed bLF in the current study has been reported in human infants. In a clinical study of preterm (;30 wk), very-low-birth– weight (
The Journal of Nutrition Nutritional Immunology
Dietary Bovine Lactoferrin Alters Mucosal and Systemic Immune Cell Responses in Neonatal Piglets1–3 Sarah S. Comstock,4 Elizabeth A. Reznikov,5 Nikhat Contractor,6 and Sharon M. Donovan4,5*
Abstract Lactoferrin (LF) is a multifunctional immune protein found at high concentrations in human milk. Herein, the effect of dietary bovine LF (bLF) on mucosal and systemic immune development was investigated. Colostrum-deprived piglets were fed formula containing 130 [control (Ctrl)], 367 (LF1), or 1300 (LF3) mg of bLF/(kg body weight d). To provide passive immunity, sow serum was provided orally during the first 36 h of life. Blood, spleen, mesenteric lymph node (MLN), and ascending colon (Asc) contents were collected on day 7 (n = 10–14/group) and day 14 (n = 10–12/group). Immune cell populations were quantified by flow cytometry and immunoglobulins (Igs) were measured by ELISA. Additionally, immune cells were isolated from spleen and MLNs (n = 7/group) on day 7 and stimulated ex vivo with phytohemagglutinin or lipopolysaccharide (LPS) 6 LF for 72 h. Secreted cytokine concentrations were quantified by multiplex assay. Lymphocyte populations [cluster determinant (CD)4, CD8, and natural killer cells] developed normally and were unaffected by dietary bLF. LF3 piglets tended to have 1.4 to 2 times more serum IgG than Ctrl piglets (P = 0.07) or LF1 piglets (P = 0.03), but IgA in Asc contents was unaffected by bLF. Asc IgA was 4 times higher on day 14 than day 7. Spleen cells from LF3 piglets produced 2 times more interleukin (IL)-10 and tumor necrosis factor (TNF)-a ex vivo than those from Ctrl or LF1 piglets. MLN cells from LF1 and LF3 piglets produced 40% more IL-10 and tended to produce 40% more IL-6 (P = 0.05) than those from Ctrl piglets. However, ex vivo bLF did not affect the cytokine response of spleen or MLN cells to LPS. In summary, dietary bLF alters the capacity of MLN and spleen immune cells to respond to stimulation, supporting a role for LF in the initiation of protective immune responses in these immunologically challenged neonates. J. Nutr. 144: 525–532, 2014.
Introduction Clinical and epidemiologic evidence suggest that human infants fed formula have altered immune responsiveness compared with those fed human milk (1,2). It is well known that human milk and formula differ in composition (3,4). One protein whose concentration markedly differs between human milk and infant formula is lactoferrin (LF)7. Human colostrum and mature milk contain ;7000 and 2500 mg LF/L, respectively (5,6). In contrast, 1
Supported by Wyeth Nutrition, a Nestle´ Business, King of Prussia, PA. Author disclosures: N. Contractor was an employee of Wyeth Nutrition at the time of this study. S. M. Donovan and S. S. Comstock have consulted for Wyeth Nutrition. E. A. Reznikov, no conflicts of interest. 3 Supplemental Methods, Supplemental Table 1, and Supplemental Figures 1–5 are available from the ‘‘Online Supporting Material’’ link in the online posting of the article and from the same link in the online table of contents at http://jn. nutrition.org. 7 Abbreviations used: Asc, ascending colon; bLF, bovine lactoferrin; BW, body weight; CD, cluster determinant; Ctrl, control, 130 mg bLF/(kg BWd); IPP, ileal PeyerÕs patch; LF, lactoferrin; LF1, 367 mg bLF/(kg BWd); LF3, 1300 mg bLF/(kg BWd); LFR, lactoferrin receptor; MLN, mesenteric lymph node; PBMC, peripheral blood mononuclear cell; PHA, phytohemagglutinin; PreLPS, LF pretreated with LPS. * To whom correspondence should be addressed. E-mail: sdonovan@illinois. edu. 2
bovine milk contains between 50 and 300 mg LF/L (7). This low concentration in bovine milk results in cow milk–based infant formulas containing small quantities of LF (8). LF is an iron-binding protein that has immune-modulating effects (9), antimicrobial functionality (10), and antioxidant activity (8). LF has direct bacteriocidal effects, but may also directly interact with the host immune system through interactions with pathogen-associated molecules, such as CpG-rich DNA or LPS (11–14). Additional immune-modulating effects of LF may be receptor-mediated as activated lymphocytes express LF receptors (15–17). As a product of neutrophil granule release, LF likely plays an important role in the local microenvironment of an immune response (18). Orally administered bovine lactoferrin (bLF) has been demonstrated to influence human immunity (19,20). Thus, milk-derived LF may be important in neonatal immune response and development. The neonatal piglet is a well-accepted model for human infant gastrointestinal function, growth, and metabolism. Sow milk contains LF (21), the LF receptors are present in the porcine intestine (22), and orally administered LF is absorbed by the neonatal piglet (23). For these reasons, we used the neonatal piglet to study how dietary bLF influences immune cell populations and
ã 2014 American Society for Nutrition. Manuscript received December 24, 2013. Initial review completed January 4, 2014. Revision accepted January 28, 2014. First published online February 19, 2014; doi:10.3945/jn.113.190264.
525
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
4 Food Science and Human Nutrition and 5Nutritional Sciences, University of Illinois, Urbana, IL; and 6Wyeth Nutrition, a Nestle´ Business, King of Prussia, PA
responsiveness in early life. We hypothesized that dietary bLF would improve the immune development of formula-fed neonates. We anticipated that bLF supplementation would increase cytotoxic T-cell numbers, prime stimulated immune cells to produce more cytokines, and increase immunoglobulin production.
Materials and Methods
Sample collection. On day 7 (n = 36) or 14 (n = 32) postpartum, piglets were sedated with an intramuscular injection of Telazol (Tiletamine HCl and Zolazepam HCl, 3.5 mg/kg BW each; Pfizer Animal Health, Fort Dodge, IA). After sedation, blood was collected by cardiac puncture into EDTA-laced vials (BD Biosciences, Franklin Lakes, NJ) for isolation of mononuclear cells. Piglets were then killed by an intravenous injection of sodium pentobarbital (72 mg/kg BW; Fatal Plus; Vortech Pharmaceuticals, Dearborn, MI) as described previously (25). Mesenteric lymph nodes (MLNs), spleens, and ileal PeyerÕs patches (IPPs) were quickly dissected and ascending colon (Asc) contents were collected. Isolation of mononuclear cells from peripheral blood and cells from immune tissues. Peripheral blood mononuclear cells (PBMCs) were isolated as previously described (26). Supplemental Methods provides detailed isolation procedures. Briefly, spleens, MLNs, and IPPs were washed. IPPs were incubated in DTT/HBSS and then with EDTA/HBSS. Spleens, MLNs, and IPPs were rinsed and single cell suspensions were made through a combination of enzymatic and mechanical digestion. Resulting cell suspensions were treated as described (25). Cell stimulation. Cells from 7-d-old piglets were plated (96-well plate, 2 3 105 cells/well) in a final volume of 200mL complete Roswell Park Memorial Institute at 37°C under 5% CO2. Stimulants were added immediately (n = 2 wells per sample per stimulant). Cells were stimulated with LPS, a toll-like receptor 4 agonist, (2 mg/mL); phytohemagglutinin (PHA), a lectin that binds and crosslinks the T-cell receptor, (2.5 mg/ml); or bLF (50 mg/mL). In some cases bLF was added with polymixin B (10 mg/ mL) to control for potential endotoxin in the bLF solution. Additionally, the following mixtures were used to stimulate cells: LPS + LF, LF + polymixin B, and LF pretreated with LPS (PreLPS) for 1 h prior to addition to cells (PreLPS + LF). Cell culture supernatants were collected 72 h after initiation of culture and frozen at 280°C until analyzed. Cytokine production. Supernatants were analyzed for IL-6, IL-10, IFNg, IL-12p40, IL-4, and TNF-a by Panomics using a Luminex-based assay 526
Comstock et al.
Phenotypic identification of cells. Cells were resuspended in flow staining buffer (PBS, 1% BSA, 0.1% sodium azide). Cell phenotypes were determined by flow cytometry as previously described (26). To differentiate T and NK cells, cells were stained with anti-pig cluster determinant (CD) 3, anti-pig CD4, and anti-pig CD8 (27). B lymphocytes were identified by anti-pig CD21 and anti-pig swine leukocyte antigen class II DR. Polymorphonuclear cells were identified by anti-pig CD163, anti-pig CD172a, and anti-pig swine workshop cluster 8 (28). Detailed information is included in Supplemental Methods. Staining was assessed by a LSRII flow cytometer (BD Biosciences) and analyzed using FlowJo 7.0 (FlowJo). Immunglobulin and bLF concentrations. Serum IgG, serum bLF, and Asc content IgA were determined using ELISA antibody sets (Bethyl Labs). The manufacturerÕs instructions were followed with the exception that PBS was used as the basis for assay buffers. Prior to analysis, Asc contents were extracted (1:3) using a PBS/protease inhibitor solution (Roche). IgA concentrations are reported as mg IgA per mg total protein in the extract. Amounts of bLF in diets were analyzed by semiquantitative Western blot at Wyeth Nutrition. Supplemental Methods provides details. Statistical analyses. Data were analyzed using SAS 9.2. Detailed information about statistical analyses is in Supplemental Methods. Proc general linear model was used in most cases with day, diet, and day diet as factors, or diet, stimulant, and diet stimulant as factors. Post hoc testing was least-squares means. For IgG, sow had an effect in the Proc general linear model, so Proc MIXED was used with fixed effects of diet and day and random effects including sow and replicate. If data were not normally distributed, they were log transformed to be normally distributed or, when log transformation failed, analyzed by nonparametric tests. LF intake was analyzed by WilcoxonÕs rank-sum test to compare days and by Kruskal-Wallis 1-factor ANOVA to compare diets. Serum bLF concentration was analyzed using a nonparametric Friedman 2-factor ANOVA. For survival analysis, the LIFETEST procedure was used to generate Kaplan-Meier survival curves and to compare survival curves among the dietary treatment groups. Differences were considered significant at P < 0.05 and trends at P < 0.10. Data are expressed as means 6 SEMs.
Results Bovine lactoferrin intake During the first 7 d, piglets consumed 103 6 13 (Ctrl), 321 6 26 (LF1), and 1170 6 39 (LF3) mg bLF/(kg BW d), respectively. Throughout the 14-d study, piglets consumed 114 6 14 (Ctrl), 413 6 27 (LF1), and 1270 6 37 (LF3) mg bLF/(kg BWd), respectively. Thus, 7-d-old and 14-d-old piglets were consuming similar amounts of bLF on a per-kg BW basis, but bLF intake differed across diets (P < 0.0001). However, on an absolute basis, piglets analyzed at day 7 consumed 256 6 25 (Ctrl), 836 6 111 (LF1), and 3130 6 201 mg bLF/d (LF3) during the 3 days prior to sampling. Piglets analyzed at day 14 consumed 496 6 82 (Ctrl), 1790 6 159 (LF1), and 5130 6 335 mg bLF/d (LF3) during the 3 d prior to sampling. Thus, the absolute dose of bLF was much higher just prior to sampling on day 14 than on day 7 within the Ctrl (P = 0.002), LF1 (P = 0.0003), and LF3 (P = 0.0002) groups. The absolute dose of LF also differed across diets (P < 0.0001). Piglet growth and health status Birth weights were similar (1.6 6 0.04 kg). Neither absolute body weight nor weight gain of piglets was affected by dietary bLF. On day 7, piglets weighed 2.5 6 0.07 kg. On day 14, piglets weighed 4.2 6 0.11 kg. Spleen weight was unaffected by dietary
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
Piglets, housing, and care. Pregnant sows (n = 10) were supplied by the Swine Research Center at the University of Illinois, UrbanaChampaign. Vaginally delivered piglets (n = 92; n = 2–17 per sow) were removed from the sow immediately, prior to consumption of colostrum, and were washed with a povidone-iodine microbicidal solution (Henry Schein Animal Health). Because piglets receive no maternally transferred antibodies prenatally, piglets were administered pregnant sow serum at birth [4 mL/kg body weight (BW)] and 5 mL/kg BW at 8 h, 22 h, and 36 h postnatal age by oro-gastric gavage to provide passive immunity. Pigs were fed a sow milk replacer formula (Animix; Supplemental Table 1) at a rate of 360 mL/(kg BW d) divided equally into 22 feedings. Piglets were assigned to 1 of 3 dietary treatment groups at birth: control formula [(Ctrl), n = 34], or formula with added bLF at either 1000 mg bLF/L [367 mg bLF/(kg BW d) (LF1), n = 29)] or 3620 mg bLF/L [1300 mg/(kg BW d) (LF3), n = 29]. Group (dietday) enrollment was balanced by sex, sow, and birth weight. The control diet contained innate amounts of 360 mg bLF/L [130 mg/(kg BW d)]. The 2 LF concentrations were selected to reflect the dose of LF consumed by breastfed human infants (LF1) or 5 times that dose (LF3) on a mg/kg BW basis. This dose assumed that human infants (;5.5 kg BW) consume about 1 L of human milk/d with an LF concentration of 1850 mg/L (24). The bLF preparation was 97% pure LF (DMV). Piglets were individually housed as described (25). All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of Illinois. Institutional and national guidelines for the care and use of animals were followed.
specific for porcine cytokines. When the value for a sample fell below the limit of detection of the assay, values were set to 0.
Bovine lactoferrin concentration in serum To determine whether dietary LF was being absorbed, bLF concentration in piglet serum was measured. Most samples were below the linear range (500–7.8 mg/L) of the assay. For piglets with quantifiable serum bLF (1 day 7 LF3, 3 day 14 LF3, and 1 day 14 LF1), the concentration was 16.1 6 3.6 mg/L serum. Because few samples contained concentrations of bLF within the linear range of the assay, absorbance values above assay background were used for statistical analysis by ranks. LF3 piglets had higher serum bLF concentrations than Ctrl or LF1 piglets (P < 0.0001, Supplemental Fig. 2). LF1 piglets had higher serum bLF concentrations than Ctrl piglets (P = 0.0001). Immunoglobulin concentrations Total serum IgG concentrations were similar on day 7 and day 14. However, LF3 piglets had higher total serum IgG than LF1 piglets (P = 0.03) and tended to have higher total serum IgG than
Ctrl piglets (P = 0.07, Fig. 1A). In contrast, diet did not affect the IgA concentrations in Asc contents, but total Asc IgA concentrations were higher on day 14 than day 7 (P < 0.0001, Fig. 1B). T-cell populations T helper cells decreased, cytotoxic T cells increased, and the CD4 to CD8 T-cell ratio decreased in PBMC, MLN, spleen, and IPP as piglets aged (Table 1). Dietary bLF did not affect T-cell populations (Supplemental Fig. 3A–C). Neither time nor diet affected the proportion of double positive CD4+CD8+ (a mature effector memory cell population in pigs) within PBMC, MLN, spleen, or IPP (Supplemental Fig. 3D). Other immune cell populations Immune cell populations were affected by age but not by dietary bLF. The NK (CD3-CD4-CD8+) cell populations in PBMC, MLN, and spleen increased, whereas PBMC and IPP B-cell populations decreased with age (Table 2). Total major histocompatibility complex II+ cell populations decreased in PBMC, MLN, and IPP, but increased in spleen, with age (Table 2). Polymorphonuclear cells in the MLN and IPP decreased with age (Table 2). Dietary bLF did not affect immune cell populations (Supplemental Fig. 4A–D). Cytokine production PHA-induced. Splenocytes from 7-d-old piglets fed 1300 mg LF/ (kg BW d) (LF3) produced more IFN-g than cells from the spleen of Ctrl piglets (Fig. 2). IFN-g production by cells isolated from the MLNs of Ctrl and LF3 piglets was similar, and dietary LF did not affect IL-6 or IL-10 production in response to PHA stimulation of cells isolated from MLNs or spleen (Supplemental Fig. 5). LPS-induced. Dietary, but not ex vivo, bLF altered cytokine production by spleen and MLN mononuclear cells. Ex vivo bLF did not affect cellular cytokine production in response to LPS stimulation (Figs. 3 and 4). This was true whether the LPS and bLF were pre-incubated for 1 hour prior to addition to cell culture (PreLPS + LF) or were added to cell culture simultaneously (LPS + LF). Dietary LF substantially affected cytokine production by MLNs (Fig. 3) and spleen (Fig. 4) cells from 7-dold piglets. MLN cells from LF1 piglets produced more IFN-g than those isolated from Ctrl piglets (Fig. 3A). MLN cells from LF1 piglets produced more IL-6 and more IL-10 than those isolated from Ctrl piglets (Fig. 3B, C). LF3 piglet MLN cells tended to produce more IL-6 (P = 0.05) and did produce more IL-10 than cells isolated from Ctrl piglets (Fig. 3B, C). MLN production of TNF-a was not affected by dietary bLF (Fig. 3D). Splenocytes from either LF1 or LF3 piglets produced more IFN-g than those from Ctrl piglets (Fig. 4A). IL-6 production by splenocytes was not affected by dietary LF (Fig. 4B). Only splenocytes from LF3 piglets produced more IL-10 or TNF-a than Ctrl piglets (Fig. 4C, D).
Discussion
FIGURE 1 Serum IgG (A) and Asc contents IgA (B) in piglets fed Ctrl, LF1, or LF3 for 7 or 14 d. Values are means 6 SEMs, n = 10–13. Asc, ascending colon; bLF, bovine lactoferrin; BW, body weight; Ctrl, control, 130 mg bLF/(kg BW d); D, day; LF1, 367 mg bLF/(kg BW d); LF3, 1300 mg bLF/(kg BW d); T, treatment.
The key finding from this study was an improvement in the immune response of colostrum-deprived neonatal piglets fed high doses of bLF, including in systemic and mucosal immune cells capable of producing higher quantities of cytokines than in piglets fed formula containing lower quantities of bLF. Although no changes in immune cell population sizes were observed, the trend toward (P = 0.07) increased serum IgG concentrations, substantially enhanced cytokine production, and lower mortality observed in piglets fed bLF support a role for LF in the initiation of protective immune responses in these immunologically challenged neonates. Dietary lactoferrin alters immune response
527
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
bLF but was affected by age, with the spleens of 14-d-old piglets (2.6 6 0.2 g/kg BW) being larger than those of 7-d-old piglets (2.0 6 0.1 g/kg BW) (P = 0.003). Liver weight (41.5 6 0.6 g/kg BW) was unaffected by dietary bLF and age. Because piglets were colostrum-deprived, there was some mortality during the study (29). Pairwise (LF3 vs. Ctrl, LF1 vs. Ctrl) comparisons were conducted. Dietary bLF at 367 mg/(kg BW d) (LF1) was not sufficient to alter survival (Supplemental Fig. 1A). However, piglets fed the highest bLF dose [1300 mg/(kg BW d), LF3] were more likely to survive than those consuming the control diet (P = 0.03) (Supplemental Fig. 1B). When survival of piglets in all 3 treatment groups was compared simultaneously, diet tended (P = 0.08) to affect survival.
TABLE 1
T-cell populations in the PBMCs, MLNs, spleen, and IPPs of 7- and 14-d-old piglets1 Day 7
P value
77.3 80.8 80.4 90.9
6 1.2 6 0.8 6 1.8 6 1.3
69.5 6 76.8 6 73.6 6 79.7 6
1.6 1.1 1.7 2.5
0.0011 0.004 0.013 0.0001
5.8 4.5 4.8 1.5
6 0.5 6 0.7 6 0.8 6 0.4
9.8 6 8.0 6 6.1 6 1.8 6
0.7 0.8 0.6 0.3
,0.0001 0.0002 0.006 0.003
19.2 40.7 46.9 314
6 2.2 6 5.3 6 6.2 6 44
9.1 6 17.1 6 20.2 6 97.3 6
1.1 4.3 3.4 15.9
,0.0001 0.0016 0.0007 0.002
5.7 11 5.9 2.3
6 0.4 61 6 0.4 6 0.2
5.7 6 11 6 5.9 6 3.6 6
0.6 1 0.3 0.5
0.89 0.71 0.96 0.14
1 T-cell populations expressed as % total CD3+ events. Values are means 6 SEMs, n = 34–38. CD, cluster determinant; IPP, ileal PeyerÕs patch; MLN, mesenteric lymph node; PBMC, peripheral blood mononuclear cell.
In previous studies, bLF was shown to increase animal growth (30–34). In our study, dietary bLF did not affect the growth of colostrum-deprived piglets. This is likely because bLF was incorporated into the diets at the time of formulation, resulting in diets that were isocaloric and isonitrogenous, whereas bLF was provided as a supplement in previous studies (32,34). The addition of high-quality protein, rather than a specific growthpromoting action of bLF, could be the reason for improved growth in some studies. Some piglets died during the study, which is to be expected because they were colostrum-deprived and not maintained in a specific pathogen-free facility. Thus, the time period between birth and gavage with sow serum, albeit short, left piglets vulnerable to potential infection. Serum IgG concentrations of piglets in this study were lower than those in piglets consuming maternal colostrum and milk (35) and were about half (36) to one-fourth the amounts (37) previously reported for colostrumdeprived piglets. These discrepancies are likely due to the specific pool of sow serum used to transfer passive immunity to the piglets in this study. Low serum Ig concentrations likely contributed to mortality. A recent study suggests that serum IgG concentrations below 10 mg/mL increase piglet mortality (29). Unlike piglets, human infants receive maternal Ig through placental transfer (38,39), with preterm infants receiving less maternal Ig (39,40). As a result, colostrum-deprived human infants are less vulnerable at birth than colostrum-deprived piglets. However, it is well recognized that even term infants are somewhat immunocompromised because of their inexperienced and underdeveloped immune systems (41–43). Additionally, the similarity of our environmental conditions to those in the real world improves the translation of this research to humans. Thus, the increased risk of piglet mortality is justified because the conditions that increase mortality improve the applicability of the research to human infants. Several additional aspects also suggest these results would apply to humans. Piglets are outbred mammals known to be excellent models for human gastrointestinal and immune devel528
Comstock et al.
opment (44–46). Specifically, T-, B-, and NK-cell development observed in the current study was similar to what is observed in human immune development (2,47–49) and what was previously observed in pigs (25). These outbred piglets were maintained in an environment containing a wide variety of microbes typical of those found in a normal room, i.e., piglets were not maintained in a specific pathogen-free environment. All piglets began the TABLE 2 Other immune cell populations in the tissues of 7and 14-d-old piglets1 Day 7 NK cells,2 CD3-CD4-CD8+ PBMC MLN Spleen IPP B-cells,3 CD21+MHCII+ PBMC MLN Spleen IPP MHCII+ cells3 PBMC MLN Spleen IPP PMN,4 SWC8+CD172a+CD1632 MLN IPP
4.2 6 1.1 6 3.5 6 0.6 6
Day 14
P value
0.8 0.2 0.9 0.1
8.8 2.1 3.9 0.5
6 1.3 6 0.3 6 0.4 6 0.1
0.0120 0.0002 0.0015 0.42
13.0 32.8 21.7 77.3
6 6 6 6
1.2 2.1 1.7 3.2
9.2 29.8 23.0 65.4
6 0.9 6 2.2 6 1.4 6 2.4
0.0320 0.33 0.53 0.0050
70.7 40.5 41.1 77.4
6 6 6 6
3.1 2.0 2.9 3.2
58.1 35.3 49.6 65.4
6 3.0 6 2.2 6 2.7 6 2.4
0.0048 0.09 0.0389 0.0086
27.5 6 2.3 28.5 6 3.1
0.0040 ,0.0001
40.1 6 3.4 54.8 6 4.5
Values are means 6 SEMs, n = 34–38. CD, cluster determinant; IPP, ileal PeyerÕs patch; MHC, major histocompatibility complex; MLN, mesenteric lymph node; PBMC, peripheral blood mononuclear cell; PMN, polymorphonuclear cell; SWC, swine workshop cluster. 2 Expressed as % CD3–events. 3 Expressed as % lymphocytes. 4 Expressed as % CD163–events. 1
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
T-helper cells, CD3+CD4+CD82 PBMC MLN Spleen IPP Cytotoxic T-cells, CD3+CD4-CD8+ PBMC MLN Spleen IPP Helper/cytotoxic ratio PBMC MLN Spleen IPP Mature effector memory T-cells, CD3+CD4+CD8+ PBMC MLN Spleen IPP
Day 14
indicates a general priming of the IL-10 response in piglets fed bLF. Because IL-10 is a key mediator of intestinal homeostasis as well as immune regulation and is expressed by both innate and adaptive immune cells (66), increased production of this cytokine
Downloaded from jn.nutrition.org at UNIV OF CALIFORNIA SAN FRANCISCO KALMANOVITZ LIBRARY on April 20, 2015
FIGURE 2 IFN-g production by PHA-stimulated spleen cells isolated from 7-d-old LF3 and Ctrl piglets. Values are means 6 SEMs, n = 4 per diet. bLF, bovine lactoferrin; BW, body weight; Ctrl, control, 130 mg bLF/(kg BW d); LF3, 1300 mg bLF/(kg BW d); PHA, phytohemagglutinin; S, stimulant; T, treatment; Ust, unstimulated.
study with normalized passive immunity achieved through oral gavage with pooled sow serum. The decreased mortality in piglets fed bLF in the current study has been reported in human infants. In a clinical study of preterm (;30 wk), very-low-birth– weight (
